D.G. Baker BNat (Hons) BEd 1,2, S.P. Myers PhD BMed ND1, I.B. Howden BCom2, E.J. Mills DPH, FRIPH3, L. Brooks PhD1, C. Avila BSc (Hons)1,2, R. Medhurst BNat4
1. Australian Centre for Complementary Medicine Education and Research, a joint venture of the University of Queensland and Southern Cross University, PO Box 157, Lismore 2480 NSW, Australia;
2. School of Natural & Complementary Medicine, Southern Cross University, NSW, Australia
3. Department of Research, The Canadian College of Naturopathic Medicine, North York, Canada;
4. Brauer Natural Medicine, Tanunda, SA, Australia
*D.G. Baker dbaker10@scu.edu.au
S.P. Myers smyers@scu.edu.au
I.B. Howden ihowden@scu.edu.au
E.J. Mills emills@mail.com
L. Brooks lbrooks@scu.edu.au
C. Avila cavila10@scu.edu.au
R. Medhurst robertm@brauer.com.au
Abstract: Reports of homeopathic clinical trials have demonstrated mixed results. Systematic reviews and meta-analyses have concluded that poor trial quality and lack of independent replication contribute to the inability to formulate a definitive conclusion concerning the efficacy of homeopathy. This research group proposes a homeopathic research platform that is scientifically rigorous and reflects homeopathic practice. The clinical trial design has application to any modality of alternative and complementary medicine that includes an interview as an integral component of clinical practice.
Background
Homeopathy is a system of medicine developed by Dr Samuel Hahnemann about two hundred years ago. It is based on the principle of Similia Similibus Curentur, (Like Cures Like), where the individual’s symptom picture is matched with the set of symptoms a homeopathic preparation is capable of causing in a healthy human being (1). Preparations are manufactured using a process of serial dilution whereby the original substance is repeatedly diluted until the dose is minute. In practice, this can go from 1 part in 100 to dilutions beyond Avogadro’s Constant (6.02252 x 1023), the point at which there is little probability of any molecules of the original substance remaining (2). Between each dilution the preparations are “succussed”, a process of repeatedly striking the preparation to agitate the solution. The combined dilution and succussion is known in homeopathy as “potentisation”. Homœopathy remains a controversial therapy due to the preparation of its dilute medications.
Rationale
The increasing demand for evidence-based medicine has been mirrored by the growth in homeopathic research over the last two decades. This research has included a variety of approaches but an emphasis on randomised controlled studies is evident. The Oxford Centre for Evidence-Based Medicine places an individual randomised controlled trial as Class 1b evidence (3). Reports of many of these studies can be found in peer-reviewed journals; however, few, if any, have been independently replicated.
Several meta-analyses have been conducted to examine the results of individual studies in a systematic manner and to test the hypothesis that the homeopathic preparation is no different to placebo. The first of these, published in the British Medical Journal (1991) (4), concluded that, despite the positive evidence of clinical trials, low methodological quality and publication bias prevented a definitive conclusion as to the efficacy of homeopathy. A second meta-analysis, published by the Homeopathic Medicine Research Group as part of the report to the European Commission in 1996 and later published in the European Journal of Clinical Pharmacology (5, 6), stated that although there is some evidence that homeopathic treatments are more effective than placebo, the overall low methodological quality of the trials, published data, and reporting prevented any firm conclusion. Similar results were demonstrated in another major meta-analysis, published in Lancet (1997) (7), which concluded that although the results were not compatible with the null hypothesis that the clinical effects of homœopathy are completely due to placebo, there was insufficient evidence to show that homeopathy was clearly efficacious for any single clinical condition. Again, low methodological quality and publication bias were held to be contributing factors. A fourth meta-analysis (8) that compared homeopathic studies with conventional medical studies concluded that small sample sizes, lack of replication, single-site research and high attrition rates have contributed to the low methodological quality of the homeopathic trials.
If applied to homeopathy, the limitations of the scientific method and of established clinical trial protocols may serve to compromise homeopathic philosophy and practice. The clinical trial protocol is particularly appropriate in the determination of cause-effect links between single treatments and specific diseases (8). However, the basic premise of classical homeopathy, like cures like, requires the prescription of a single therapeutic substance that matches the totality of symptoms present in the individual. Homeopathy focuses on stimulating broad healing processes that influence a variety of conditions and symptom patterns (8). In the formulation of protocols for homeopathic research, this premise is often ignored because it does not fit easily into the conventional clinical trial method. As a result, the protocols of many homeopathic clinical trials have mirrored those of pharmaceutical trials, i.e. they test the efficacy of a single homeopathic preparation in a specific disease. The results of these studies are often inappropriately generalised to homeopathic practice.
A common misunderstanding of homeopathy restricts it to the use of ultra-dilute therapeutic substances. However, this denies the role of the homeopathic consultation, an integral part of the process without which the individualisation of prescriptions is impossible. The dispensing of a single homeopathic preparation across a group according to a diagnosed illness negates the need for a consultation and does not reflect homeopathic philosophy or practice.
The diverse methods of homeopathic trials, the difficulties associated with achieving congruence between trial method and homeopathic philosophy and practice, and the lack of replication of results contribute to a generally poor perception of homeopathic research. It is the contention of this research team that an adequate research method for homeopathic clinical trials has not yet been formulated. Our aim has been the development of a research platform that embodies the principles and practice of homeopathy whilst at the same time retaining scientific rigour. The platform is designed to be adapted not only for many different research questions of homeopathy, but also to the wider field of naturopathic research.
Methods/design
The platform utilises the randomised, controlled clinical trial method to enable the comparison of homeopathic treatment with and without a homeopathic interview, as well as the individualised homeopathic prescription with a generic homeopathic complex and a placebo. Figure 1 presents the research platform in flow chart format, detailing the randomisation of the study population into five arms.
Figure 1: Flow chart of the research platform
Randomisation 1:
The initial randomisation allocates subjects to one of two cohorts:
- Those who will receive a homeopathic interview, and
- Those who receive no interview.
This step is a major component of the protocol as it allows the effect of the consultation to be measured.
Randomisation 2:
The cohort not receiving an interview is randomised to receive either a placebo or a generic homeopathic preparation. This generic preparation can be either specially manufactured for the trial or it may already be commercially available as an ‘over the counter’ item. The growth in usage of such preparations reflects the growth in self-prescription and usage of natural health products around the world.
Randomisation 3:
Subjects are randomised to receive the individualised homeopathic treatment, the generic homeopathic preparation, or the placebo.
Administration:
Administration of the study preparations utilises the “double dummy” method whereby all subjects receive two preparations; the individualised preparation (or its placebo) and the generic preparation (or its placebo). No subject will receive two active preparations, however, two groups will receive both placebo preparations. Subsequent to the prescription process, all preparations are posted to the subjects by an independent Study Dispenser to maintain the double blinding protocol. Table 1 illustrates the administration schedule subsequent to the individualised intervention.
| INDIVIDUAL INTERVENTION |
SUBSEQUENT INTERVENTION |
|
|
Complex |
Placebo |
|
| Active | X |
P |
| Placebo
|
P |
X |
|
X |
P |
|
Table 1: Intervention schedule subsequent to the individualised prescription administration
Study population:
The research platform allows for subject recruitment according to a pre-determined diagnosis which facilitates the selection of primary and secondary outcome measures. While classical homeopathic practice is not dependent upon a diagnosis, it is included in the study design to assist in the recruitment of a more homogenous sample and thus to increase the internal validity of the trial. The elimination of possible uncontrolled variables enhances scientific rigour.
The Homeopathic Interview:
It is proposed that a team of two experienced homeopaths will jointly conduct the interview, which will be videotaped. As a result of this interview, all subjects will be prescribed an individual homeopathic preparation, although only one of the allocated groups will receive the active preparation. Each homeopath will independently arrive at a prescription for each subject with the final prescription being reached by consensus. In the case where consensus is not possible, two options are available:
(a) An independent third homeopath may be called upon to review the video-recorded interview and, in consultation with the original team, make a decision on the prescription, or
(b) The subject may be removed from the study. However, this course of action may cause the study population to be skewed and be regarded as a potential source of bias.
Summary
The four randomisation points result in five arms to the study:
1. Generic homeopathic complex without associated interview
2. Placebo without interview
3. Individualised homeopathic simplex with interview
4. Generic homeopathic complex with interview
5. Placebo with interview
The randomisation of subjects to interview/non-interview cohorts ensures that, given the worst case scenario where the interventions demonstrate no difference from placebo, it is at least possible to measure the effect of the homeopathic interview process.
Discussion
The proposed research platform is designed to investigate whether homeopathy has an effect greater than placebo. Despite the inclusion of the interview as part of the homeopathic process, this design does not reflect the complete process of homeopathic practice, but only the initial prescription and observation period. In practice, the prescription may actually be changed according to subsequent homeopathic assessment(s) and changes in symptoms. Homeopathic treatment of chronic conditions can be likened to the peeling of successive onion layers where the treatment of one set of symptoms may unmask underlying symptoms. This involves a different research question for which a different study method may need to be considered. The n of 1 trial, which is designed to determine the optimal treatment for the individual patient in the clinical setting, may be more appropriate for such research.
Generic homeopathic complexes usually include a group of substances that, individually, may show efficacy in any specific condition. The inclusion of a set of such substances in a single preparation is an attempt to provide a product that will suit a wider set of people than would an individual homeopathic preparation. The resulting complex preparation is thus able to be marketed as an over the counter (OTC) item that does not necessarily require a homeopathic interview. An important difference between the individualised prescription and the complex formulation is the level of potency of the constituents. Individual substances in a complex formulation are usually included at potencies that are below (i.e. less dilute than) that which would normally be administered by a classical homeopathic practitioner.
The possibility exists that some of the group allocated to the complex treatment may receive a substance that, had they been randomised to the individualised treatment group, they would have received as a simplex preparation. For this sub-group the difference in treatment effect between the complex and the individualised preparation may potentially be decreased. A comparison of the differences between treatment groups that both include and exclude this sub-group from the complex group should reveal any significant effect if one exists.
As a corollary to this, the blinding of the prescribing homeopaths to the complex constituents is of great importance. This should ensure that their decision in the selection of the individualised preparation is not biased by knowledge of the complex formulation.
In this era of evidence-based medicine, the acceptance of homeopathy relies upon a sound research foundation. It is no longer acceptable to use the myriad of case studies and observations as the sole source of evidence. The recommendations contained in the Kleijnan et al., Cucherat et al., and Linde et al. meta-analyses emphasise the need for more rigorous research to be conducted. This novel approach to homeopathic research has been developed in response to these demands. The protocols combine the rigour of the randomised controlled trial with classical homeopathic practice and principles.
Authors’ contributions:
DB conceived of the study, participated in its design and was the primary author of the manuscript.
SM conceived of the study, participated in its design, was the research methodology consultant and primary editor of the manuscript.
IH participated in the design of the study, was the primary homeopathic consultant, and assisted in the editing of the manuscript.
EM participated in the design of the study and acted as a consultant research methodologist.
LB participated in the design of the study and was the primary statistician.
CA participated in the design of the study and acted as a homeopathic consultant.
RM participated in the design of the study.
References
1. Hahnemann S. Organon of Medicine. 6th ed. New Delhi: B. Jain; 1997.
2. Atkins P, Jones L. Chemistry: Molecules and Matter. 3rd ed. New York: WH Freeman and Co; 1997.
3. Phillips R, Ball C, Sackett D, Badenoch D, Straus S, Haynes B, et al. Levels of Evidence and Grades of Recommendation. Centre for Evidence-Based Medicine, Internet, http://www.cebm.net/levels_of_evidence.asp,04/07/2003
4. Kleijnan J, Knipschild P, ter Riet G. Clinical Trials of Homeopathy. British Medical Journal 1991;302:316-23.
5. Boissel JP, Cucherat M, Haugh MC, Gauthier E. Critical Literature Review on the Effectiveness of Homeopathy: Overview of Data from Homeopathic Medicine Trials. Brussels: Commission of the European Communities, Directorate-General XII. December 1996.195-210.
6. Cucherat M, Haugh MC, Gooch M, Boissel JP. Evidence of clinical efficacy of homeopathy. A meta-analysis of clinical trials. HMRAG. Homeopathic Medicines Research Advisory Group. European Journal of Clinical Pharmacology 2000;56(1):27-33.
7. Linde K, Clausius N, Ramirez G, Melchart D, Eitel F, Hedges L, et al. Are the clinical effects of homeopathy placebo effects? A meta-analysis of placebo-controlled trials. The Lancet 1997;350(9081):834-44.
8. Jonas WB, Anderson R, Crawford C, Lyons J. A systematic review of the quality of homeopathic clincical trials. BMC Complementary and Alternative Medicine 2001;1(12).